CaGIS Vol. 29, no. 1

Visual Search for Land Use Objects in Aerial Photographs

Robert Lloyd and Michael E. Hodgson

Although visual search has been a topic of interest for some time, most search processes have considered simple target symbols in a plain background. The purpose of the current study was to consider more realistic search processes that involved complex targets and backgrounds. A cognitive experiment was conducted in which subjects searched for specific types of land use in aerial photographs. Subjects were presented displays with simple or complex backgrounds that had from one to five primary and secondary objects related to four different types of land use. Primary objects were stores, schools, churches, and farmhouses; secondary objects included parking lots, athletic fields, cemeteries, and agricultural fields. Subjects searched each photograph and determined if the stated target was present or absent. The efficiency of the visual search was measured by reaction time. Reaction times significantly decreased as the number of objects in the photograph increased. The separation of primary and secondary land-use objects significantly increased reaction times. Searches done with complex backgrounds were significantly longer than searches done with simple backgrounds. Visual searches were significantly faster when schools and churches were targets in the photographs and significantly slower when stores and farmhouses were targets.

Keywords: visual search, aerial photographs, reaction times

Qibla, and Related, Map Projections

Waldo Tobler

The qibla problem determination of the direction to Mecca has given rise to retro-azimuthal map projections, an interesting, albeit unusual and little known, class of map projections. Principal contributors to this subject were Craig and Hammer, both writing in 1910. A property of retro-azimuthal projections is that the parallels are bent downwards towards the equator. The resulting maps, when extended to the entire world, thus must overlap themselves. An unusual recent discovery from Iran suggests that Muslims might have been prior inventors of a similar projection, by at least several centuries. A later corollary by Schoy leads to a new “cylindrical” azimuthal map projection with parallels bending away from the equator, here illustrated for the first time.

Keywords: Azimuthal directions, map projections, Mecca, qibla, retro-azimuthals

An Evaluation of Fractal Methods for Characterizing Image Complexity

Nina Siu-Ngan Lam, Hong-lie Qiu, Dale A. Quattrochi and Charles W. Emerson

Previously, we developed an integrated software package called ICAMS (Image Characterization and Modeling System) to provide specialized spatial analytical functions for interpreting remote sensing data. This paper evaluates three fractal dimension measurement methods that have been implemented in ICAMS: isarithm, variogram, and a modified version of triangular prism. To provide insights into how the fractal methods compare with conventional spatial techniques in measuring landscape complexity, the performance of two spatial autocorrelation methods, Moran’s I and Geary’s C, is also evaluated. Results from analyzing 25 simulated surfaces having known fractal dimensions show that both the isarithm and triangular prism methods can accurately measure a range of fractal surfaces. The triangular prism method is most accurate at estimating the fractal dimension of surfaces having higher spatial complexity, but it is sensitive to contrast stretching. The variogram method is a comparatively poor estimator for all surfaces, particularly those with high fractal dimensions. As with the fractal techniques, spatial autocorrelation techniques have been found to be useful for measuring complex images, but not images with low dimensionality. Fractal measurement methods, as well as spatial autocorrelation techniques, can be applied directly to unclassified images and could serve as a tool for change detection and data mining.

Keywords: Fractal measurement, spatial autocorrelation, simulated surfaces, data mining

An Inverse Solution to the Winkel Tripel Projection Using Partial Derivatives

Cengizhan Ipbuker

In cartographic applications it is frequently necessary to transform the rectangular coordinates from one projection into another. In this case, one must first calculate the geographical coordinates from the rectangular coordinates of the existing map and then project these new geographical coordinates to the desired projection. This is called an inverse solution. If both of the plane coordinates are functions of the variables longitude and latitude, it may not be easy to derive the geographical coordinates. This paper describes an iterative approach for the inverse solution of the Winkel Tripel projection using partial derivatives. I chose the Winkel Tripel projection because it is commonly used for mapping the whole world. It has a special importance in atlas cartography where it is regarded as a suitable projection with relatively little distortion, distributed more uniformly than many other atlas projections.

Keywords: Winkel Tripel, inverse solution, Newton’s iteration, partial derivatives, Jacobian matrix

Towards a Participatory GIS: Evaluating Case Studies of Participatory Rural Appraisal and GIS in the Developing World

Brian H. King

The recognition that local participation is a critical goal of development has contributed to the popularity in a set of techniques designed to increase local participation and knowledge in planning processes. Identified as participatory rural appraisal (PRA), this trend is marked by the use of a variety of high-end technologies, including geographic information systems (GIS). An interesting and related trend has come from members of the GIS community who argue that a “participatory GIS” is required to ensure local knowledge and participation in a variety of planning initiatives. This synergy of interests has resulted in a growth of research in the developing world that attempts to merge PRA methods with GIS tools. This paper examines the separate, but increasingly complementary, traditions of PRA and GIS. Ten case studies that combine participatory methods with GIS in Latin America, Africa, and Asia are evaluated to consider how these traditions are being applied by communities to protect ownership of territory, present local knowledge of natural resources, and to engage in long-term planning. This paper suggests that although GIS has the potential to increase participation in planning processes, a commonality of the majority of case studies is limited attention to the participatory process itself. It would address how access to information and GIS tools varies within communities, as well as the effectiveness of a participatory GIS in shaping policy outcomes. Although participation can be advanced through the use of GIS, this paper concludes by suggesting that research on the availability of these tools should not serve as a substitute for critical analyses of their use and effectiveness by local communities.

Keywords: Participatory GIS, community-based GIS, GIS and society, PRA, development

Book Review

Geographic Information Systems and Science, Longley P A, Goodchild M F, Maguire D J, and Rhind D W (eds). John Wiley & Sons, Chichester, Sussex, 2001. 454 pages. $87.56 cloth (ISBN 0-471-49521-2).

Geographic information systems (GI systems) and science (GI science) have been progressing at blinding speeds as geographic information scientists (GI scientists) continue the quest for new ways of describing, understanding, and predicting the spatial interactions between humans and the earth. However, in this progress a disproportionate amount of published materials focus on the spatially aware professionals (SAPs). There are only a few books that we can truly say provide a holistic coverage of the GIS discipline for non-SAP audiences. For many years, Burrough’s Principles of Geographical Information Systems for Land Resources Assessment was the main textbook available. Burrough’s book, coupled with the authoritative two-volume Geographical Information Systems: Principles and Applications, were the staple of many university introductory GIS courses. Stand-alone textbooks with geographic information education goals are scarce.

It is in this context that Geographic Information Systems and Science by Longley et al. is a much needed addition. In their book, the authors use a broad brush to paint a comprehensive picture of the GIS discipline in a clear and non-threatening style of writing, showing trans-disciplinary relationships where they exist, providing a copious supply of examples to reinforce theory, highlighting influential personalities and their work, and pointing the way to Internet-based materials for readers who wish to explore topics in more detail.

Geographic Information Systems and Science has a foreword and afterword by Joe Lobley, a preface, a list of acronyms, an epilog, and nineteen chapters covering the principles (ideas), techniques (concepts) and practice (action) of geographic information problem solving. The rapid trans-disciplinary growth of GIS, the differing learning requirements for SAPs and non-SAPs, together with changing education modes and the increasing influence of information and communication (ICT) tools in the learning process have all shaped the form and content of this book. Improving capability to address geographic problems through better understanding, improved expertise, and recognition of unintended consequences and strong professional values are the main topics of this work, as the authors point out in the preface.

In the introduction section consisting of chapters 1 and 2, GIS is placed in a problem-solving context. The history of GIS, issues of scale, problem-solving paradigms, and the content of GI systems, GI science, and GI education are gradually outlined. This is supported by a gallery of applications that emphasise the spatial nature of our everyday lives. For the new reader, the applications provide a real-world experience map to aid personal navigation through the theory that follows. Chapters 3 to 7 unmask the scientific principles that govern the discipline. Theoretical issues such as representing geographic space and processes, georeferencing, characteristics of geographic data, errors and uncertainty in representing reality, and the important role of metadata in spatial data interchange are discussed in conceptual rather than technical detail. Chapters 8 to 15 form the second section of the book and they cover the main techniques of GIS analysis. Starting with a broad-based discussion of commercial GIS software, the content progresses through the chapters using the traditional data gathering, data input and processing, spatial data analysis and visualization, and error analysis process flows that characterize GIS data handling. By showing the links between GIS applications, scientific principles, and people-oriented empirical analysis, the authors set the scene for the next section. Chapters 16 to 19 comprise the final section that is dedicated to examining the role of people and organizations in GIS practice. How do we integrate people and GIS technology in a holistic manner to effectively address the objectives of the organization? To understand this question, the authors present an organizational analysis perspective that casts a broad net on issues such as operational planning and management, asset and risk management, and collaboration strategies between public and private organizations at local, regional, and global levels. The epilog summarizes the chapters and places them in proper perspective. After reaching this stage of the book, the new reader is better able to understand the current status of GI systems, science, and studies, and to critically follow the authors on a short journey as they look ahead to the potentials and challenges of GIS in the coming years.

Overall, this is an excellent book for readers looking for a user-friendly introduction to the links and concepts of GIS. It is a textbook, and strong pedagogical features predominate to give readers refreshing ways of interacting with the content. At 10” x 7.5” x 1” and full-color, it grabs your attention immediately. A further look under the cover reveals that each chapter starts with an overview and clear learning objectives, and that there are text boxes to highlight important concepts or to present personality profiles. There is a strong focus on geographic problem solving through examples and sufficient in-depth theory to aid understanding and to chart solution strategies. Questions for further study, general online Internet resources, reference links to mostly its technical predecessor, Geographical Information Systems: Principles and Applications, and links to the ESRI Virtual Campus and the NCGIA Core Curricula round off the diversity of quality interactive content that is packaged within the pages of this book.

However, if there is one shortcoming it has to be the neglect of any serious discussion about freeware and shareware GIS. In some developing countries, and for many individuals, the present price of commercial GIS software is a barrier to widespread adoption of the technology. For this sector of the digital divide, freeware and shareware GIS software provide sometimes the only means for practical GIS training and education. The social and political consequences of this are of significance and deserve some mention in the book. Nevertheless, this missed opportunity does not diminish the tremendous effort of the authors and the value of this work as the most complete scholarship that presently exists on geographic information science.

This book will be of great value to undergraduate and graduate students studying GIS, serious lay readers interested in the subject, and GIS managers and academics who may need to refresh their GIS knowledge. [Shivanand Balram and Suzana Dragievic, The Spatial Analysis and Modelling Research Group, Simon Fraser University]

Book Review

Parallel Processing Algorithms for GIS, Richard Healey, Steve Dowers, Bruce Gittings, and Mike Mineter (eds). Taylor and Francis Ltd., London. 1997. 470 pages. Hardcopy $105 (paperback $48).

The increasing availability of large quantity of GIS and remote sensing data, the growing complexity of spatial algorithms, and the burdensome requirement of real-time GIS operation are the most compelling reasons for exploiting parallel computation in the GIS context. In the past twenty years, a great deal of progress in parallel GIS and remote sensing has been made worldwide. However, most of those results are limited in algorithmatical level or to a particular parallel computer platform. Parallel GIS and remote sensing has not grown at a healthy and speedy pace due, in part, to a lack of a systematic study as well as a good textbook on the subject. Parallel Processing Algorithms for GIS, edited by Richard Healey, Steve Dowers, Bruce Gittings, and Mike Mineter, and published by Taylor & Francis, fills this gap and is indeed a groundbreaking book on this subject.

Unlike other edited collections, which usually lack a sustained goal, this book aims to give readers “a clear picture of both the technical problem and the benefit/cost of developing GIS algorithms in a parallel environment.” It does a very nice job in meeting its goal. The editors as well as the authors of a large portion of this book are from the University of Edinburgh, a leading parallel GIS research institute in the world. Their study on parallel GIS is much more serious, comprehensive and systematic than any other places, which is reflected by the content of this book.

This book has five parts. Part 1 discusses basic concepts, terminology, and techniques of parallel processing. The readers should be able to grasp the fundamental issues, such as hardware, software, basic parallel programming concepts, and data I/O. Parts Two and Three originate from a research project on parallel GIS at the University of Edinburgh. Part Two serves as a cornerstone of the whole book and addresses the general issues of parallel software design in a GIS application, including data structure (vector and raster), data format and topology, data management in a parallel environment and software reusability. Part Three gives some examples of developing parallel algorithms. Three algorithms are described: vector-raster conversion, raster-vector conversion, and polygon overlay. Part Four presents research progress achieved elsewhere.

In the past twenty years, people’s perspective about parallel computation has changed from overly optimistic and enthusiastic to more realistic and practical. A proper review about parallel GIS is given in Part Five. Although this is the shortest part, it presents some very important issues, such as hardware, software, data, and expertise, which are too important to be ignored by any serious parallel GIS professional.

Although this book gives a good systematic study on parallel GIS, it is apparent that parallel GIS as well as GIS is still lacking in theoretical fundamentals. Currently, the most successful and active areas of parallel computation are physics, climate modeling, material science, and biology. One common reason for their successes is that all of them have very solid theoretical fundamentals. In a typical parallel computation project, good theories and models usually help to set up the project objective, guide the algorithm development, and interpret the result. Unfortunately, many parallel GIS projects lack such kind of theoretical support and therefore their objectives are merely to speed up the operations, which is much less appealing and valuable. This is the area that neither geography nor this book has any clear answer for at this moment.

Finally, I would like to reiterate a point from this book to end this review. “There is a shortage of skilled staff in parallel processing, there are even fewer individuals with expertise in the complex and sometime arcane world of GIS algorithm. To find those with both types of skills is therefore even more difficult, and makes interdisciplinary collaboration essential”(p.4). This book builds a nice bridge between the GIS and parallel computation communities. If you are an established GIS professional, this book might echo many of your experiences, thoughts, and concerns from different viewpoints. If you are a parallel computer expert, this book could help you to enter a new and very promising territory. If you are an enthusiastic parallel GIS newcomer, this is an excellent reference book and a guideline. [Yang Cheng, Jet Propulsion Laboratory, Pasadena, California]